The Public Is a Thermostat

My colleague and co-author Matt Grossmann suggested this post.

David Brooks sees the public as largely opposed to the policies of the Obama administration and the Democratic majorities in Congress. He believes that this reflects a miscalculation on the Democrats’ part: the public is not that liberal.

Some Kool-Aid sippers on the left say the problem is that Republicans have better messaging (somehow John Boehner became magically charismatic to independents). Others say the shift to the right is a product of bad economic times. But Dr. Faustus saw a deeper truth. Moderate suburban voters do not see the world as liberals do, even in the most propitious circumstances, and never will.

But there is another possibility: the public is simply a thermostat. When government spending and activism increases, the public says “too hot” and demands less. When spending and activism decreases, the public says “too cold” demands more. Here is Christopher Wlezien in a 1995 paper (gated):

We observe that the signals the public sends to policymakers, in the form of preferences for “more” or “less” spending, react to changes in policy…[T]here is negative feedback of spending decisions on the public’s relative preferences, whereby the public adjusts its preferences for more spending downward when appropriations increase, and vice versa.

Erikson, MacKuen, and Stimson, writing in chapter 9 of The Macro Polity, refer to “the governing system as a thermostat.” Erikson et al. show that the public’s “mood”—a general measure of the policies it desires—moves in the opposite way as policy:

The correlation between policy innovation in one administration and before-after mood change is a strongly negative -0.76…The more liberal the policy stream, the more conservative is the change in mood. Notably, the most liberal presidency (Johnson’s full term ending in 1968) is associated with the greatest public reaction in the conservative direction. Similarly, the conservative presidencies of Reagan and Eisenhower moved the public in a liberal direction.

Brooks is wrong to assume that the public’s reaction to Democratic policies indicates a enduring ideological disjuncture or a failure of public relations. The public may not be more conservative. It may simply be saying “too hot.” As Matt put it in his email to me:

Current trends would not show that Democrats have been unusually unsuccessful in moving public opinion but that policy ideology in public opinion typically moves against the direction of policymaking. The public requests liberal policies, gets them, and then moves in the other direction; they then get more conservative policies and move against them.

Brooks wants to score this moment as a victory or defeat for someone—in this case, a defeat for liberalism and the Democrats. But If policy and thermostatic public opinion is cyclical, then any victory or defeat is temporary. The ebb and flow is the more important dynamic.

10 Responses to The Public Is a Thermostat

  1. Benjamin Geer June 22, 2010 at 7:18 pm #

    But can you explain why this would happen? And can you square it with individuals’ perceptions of their own political positions? I haven’t changed my political views on most issues in decades. I certainly don’t change them in response to government policies. I’ve always wanted more liberal policies, regardless of which policies were implemented.

  2. John June 23, 2010 at 5:26 am #

    It’s about the groups being mobilized. As liberal policies are enacted it creates a backlash among conservatives who seriously object to the policies. Same with conservative policies sparking liberal outrage. Unhappy people are more likely to vote and mobilize because they aren’t getting their way. Since not everyone votes those groups have a disproportionate impact on policy. You’re less likely to be interested in politics if you are happy with how things are going. Most people aren’t shifting their positions so much as shifting their level of political participation.

  3. Stan Olshefski June 23, 2010 at 9:44 am #

    I honestly would love to see the spend less theory tested at the federal level.

    Seeing that federal spending has never been in the post-war period, how did you come to that hypothesis?

  4. John Sides June 23, 2010 at 10:54 am #

    Stan: I didn’t come to any hypothesis. I’m describing the work of other political scientists. But to answer your question: you look at specific areas of spending that have gone up and down, such as defense spending, and the public’s preferences about spending in those areas.

  5. Matt June 23, 2010 at 11:01 am #

    I agree with John that mobilization is a good explanation for the changes in election outcomes, but this would not explain changes in average policy opinions of all citizens. This change would be expected by The Macro Polity authors, but I have not seen enough evidence to know whether it is happening now.

    If policy opinions are moving in a conservative direction, I see two explanations. First, the public is moderate. Median public policy preferences are between the preferences of Democrats in Congress and Republicans in Congress. When either party gets their way, they are pursuing policy to the left or right of voters. This is Brooks’ point, though he incorrectly believes it is restricted to Democrats. Second, the public does not have well formed opinions about their ideal points on public policy. Instead, they have a general idea about whether the current moment requires more government services and regulation or not. In 2008, they had heard a lot about the need for more; now, they have heard a lot about the need for less.

  6. Jim Shoch June 23, 2010 at 11:34 am #

    I think Matt’s comment is right on. The public, or median voter, is essentially moderate on spending issues. Parties in power, pressured by their core constituencies, regularly overshoot the median voter, leading to the “thermostatic” reaction John describes. Check out Chris Wlezien and Stuart Soroka’s new book, “Degrees of Democracy.” Dan Wood makes essentially the same argument with respect to presidential overshooting in his own book, “The Myth of Presidential Representation.”


  1. No, the 2012 election didn’t prove the Republican Party needs a reboot - March 31, 2013

    […] serve as “thermostats” for public policy.  They move in the opposite direction as the party controlling the White […]

  2. hot models » No, the 2012 election didn’t prove the Republican Party needs a reboot - March 31, 2013

    […] offer as “thermostats” for open policy.  They pierce in a opposite instruction as a celebration determining a White […]

  3. hot models » The GOP is just fine - March 31, 2013

    […] offer as “thermostats” for open policy.  They pierce in a opposite instruction as a celebration determining a White […]

  4. Toename euroscepsis vooral bij lageropgeleiden | Sargasso - June 15, 2013

    […] door de politieke en economische context waaronder daadwerkelijke beleidsmaatregelen (zie hier en hier). Om het huidige eurosceptische tij te keren zullen beleidsmakers dus vooral naar […]